As used herein and in parent U.S. patent application Ser. Nos. 13/088,059 and 12/802,447, “sulfation” or “sulfonation” is the post-translational addition of a sulfate moiety to a protein.
Although the nature of a protein is dictated primarily by the particular amino acid sequences derived from transcription of its nucleic acid coding sequence, there are post-transcriptional processes that may also affect its properties. Some of these modifications are large scale rearrangements such as: (a) conversion of an inactive pro-enzyme into an active form by removal of part of an amino acid sequence; (b) protease digestion of a composite protein into individual segments with varied functions as seen in some viral proteins (for instance, the polyprotein of HIV); or (c) removal of an internal amino acid sequence (an intein) by protein splicing. In addition to these cleavage processes, modification of individual amino acids can take place by enzymatic addition of functional groups such as methyl, acetyl, phosphate, glycosyl, palmitoyl, sulfate and ubiquitin groups.
The difference in functionality caused by these modifications can induce radical differences in properties. For example, proinsulin is an inactive enzyme that is only found in its active form (insulin) after proteolytic cleavage transforms the protein into separate peptide chains connected by disulfide bonds. In another instance, the addition of a ubiquitin moiety does not necessarily affect its enzymatic functions but generates a signal for degradation of the “tagged” protein. Even relatively modest alterations, such as acetylation and phosphorylation of one or more amino acids in a protein, can induce remarkable changes in the properties of a protein target. The importance of both of these processes in controlling levels of activities within cells by such modifications can be seen by the abundance of substrate specific versions of each of these family of proteins (acetylases and kinases) within a cell. Further control is exerted by the action of proteins that reverse these changes, i.e., de-acetylases and phosphatases. These modifications can result in an increase or a decrease in the activity level of the target protein and/or a change in its physical locale.
Although the kinase and acetylase modifications are well known areas of research, the importance of sulfation is receiving increased attention. For recent reviews see Stone et al., 2009 New Biotechnology 25:299-317 and Monigatti et al., 2006 Biochim Biophys Acta 1764:1904-1913. Sulfation of tyrosines is believed to take place in about 1% of the tyrosines in proteins and appears to facilitate protein-protein interactions (Baeuerle and Huttner 1985 JBC 260:6434-6439, Kehoe and Bertozzi 2000 Chem Biol 7:R57-R61). Of particular interest is the connection between sulfation with receptors and their ligands, since the enzymes responsible for sulfation, tyrosylprotein sulfotransferase-1 (TPST1) and TPST2, are localized in the Golgi apparatus.
Although sulfation has been mostly studied in cytokine receptors and their ligands, it has been recently noted that unsulfated Wnt does not generate as strong a signal as sulfated Wnt, presumably due to a differential ability of the unsulfated ligands to bind the LRP5/6 receptors that are involved in the Wnt signaling system (Cha et al., 2009 Current Biol 19:1573-1580). In addition to tyrosine, evidence has become available that serine and threonine are also potential sites, although at the present time it is not known if this is carried out by the same enzymes that modify tyrosines (TPST-1 and TPST-2) or if another enzyme or enzymes are responsible (Medzihradszky et al., 2004 Molec Cell Proteomics 3:429-440). The increased binding of sulfated proteins for their binding partner is, at least in some cases, due to the formation of a salt bridge between the sulfate group and arginine residues on the binding (see Woods et al., 2007, J. Proteome Res. 6:1176-1182 and references cited therein).
Testing for the presence of sulfation modifications in a protein can be carried out using various methods (for reviews, see Monigatti et al. 2006, Stone et al. 2009, and Seibert and Sakmar 2007 Polymer 90:459-477). The two most popular methods for this type of analysis is the use of mass spectrometry (MS), or antibodies that are specific for Sulfo-Tyr. With regard, to mass spectrometry, definitive answers on the presence of sulfated tyrosines can be achieved, but due to the lability of the bond between the sulfate group and tyrosine, special modifications have to be made to the standard mass spectrometry protocols (Drake and Hortin, 2010 Int J Biochem Cell Biol 42:174-179). In a more biological approach, antibodies have been developed that can detect the presence of sulfated tyrosine residues. Antibodies have been developed that can detect the presence of sulfated tyrosines regardless of the particular peptide sequence they are embedded within (Kehoe et al., 2006 Molec Cell Proteomics 5:2350-2363; Hoffhines et al., 2006 J. Biol Chem 281:37,877-37,887). The general nature of their recognition allows a wide variety of different proteins to be identified as long as they contain a sulfated tyrosine. In many cases, proteins have to be isolated or separated for this type of analysis to observe individual effects, since there is no discrimination between the different sulfated proteins by such antibodies. For instance, the extent of sulfation can be determined for individual isolated proteins of interest or patterns of a group of proteins can be analyzed. In an alternative approach, antibodies have been developed for specific proteins with a sulfated tyrosine. These antibodies can detect differences between sulfated and non-sulfated forms and can identify the presence of the sulfated protein in a mixture of other proteins (Bundgaard et al., 2008 Methods Mol Bio 446:47-66). The specificity of the epitope requires that a new antibody has to be developed for each particular protein of interest.
As information has accumulated concerning the amino acid sequences that are used as substrates for sulfation, it has become clear that there is no simple consistent recognition sequence (see, e.g., Niehrs et al., 1990 JBC 265:8525-8532, Bundgaard et al., 1997 JBC 272:31,700-31,705). A computer program called “Sulfinator” has been created recently that is capable of analyzing protein sequences and predicting the presence or absence of sulfation sites (Monigatti et al. 2002 Bioinformatics 18:769-770). The program achieves its highest accuracy only when proteins are tested that are either receptors, or ligands for receptors, because these are proteins that are processed through the Golgi apparatus where the TPST-1 and TPST-2 enzymes are localized. Proteins that are cytosolic in nature are physiologically irrelevant since even if they have appropriate sequences they would never come into contact with the tyrosine sulfotransferases. The Sulfinator does not detect the extent of sulfation.
In detecting the extent of sulfation, experiments have shown that even proteins that are substrates for sulfation do not always represent a homogeneous population with complete sulfation. For example, gastrin peptides, which are easily sulfated, show a mixed population of both sulfated and unsulfated forms in roughly equal proportions (Hilsted and Rehnfeld 1987 JBC 262:16,953-16,957). In another instance, there may be tissue specific differentiation on the extent of tyrosine sulfation of Chromogranin A that depends upon whether it is made in parathyroid or adrenal cells (Gorr and Cohn, 1999, JBC 265:3012-3016). Different effects have also been observed for proteins such as gastrin/cholecystokinin peptides and their precursors where varying degrees of modification are seen during ontogenesis and pathogenesis of certain diseases (Rehfeld et al., 1989 Biochimie 70:25-31). Furthermore, in certain circumstances, such as in the expression of cloned recombinant proteins, there may be undersulfation of proteins that would otherwise be completely modified (Seibert and Sakmar 2008 Biopolymers 90:459-477).
Although extensive efforts have been made in searching for pharmaceutical agents that affect kinase activity, compounds that affect sulfation modifications have only recently attracted attention (see, e.g., Hemmerich et al., 2004 Drug Discovery Today 9:967-975). The potential utility of influencing sulfation reactions can be seen, however, by recent discoveries that CCR5, one of the receptors for recognition of HIV, is sulfated. The importance of this modification can be seen by results with chlorate (an inhibitor of tyrosine sulfation), where the presence of this factor decreases the affinity of gp120/CD4 complexes toward the CCR5 receptor (Farzan et al., 1999 Cell 96:667-676). Although there are instances where the presence of a sulfation modification enhances binding, there are also numerous instances where there is an absolute requirement for sulfation to take place in order for certain proteins to have biological activity (Farzan et al., 2001 J Exp Med 193:1059-1065; Costaglia et al. 2002 EMBO J 21:504-513; Gao et al., 2003 JBC 278:37902-37908; Gutierrez et al., 2004 JBC 279:14726-14733; Hirata et al., 2004 JBC 279:51775-51782, Fieger et al., 2005 FASEB J 19:1926-1928 and Colvin et al., 2006 Molec Cell Biol 26:5838-5849).
Furthermore, in vitro studies also show the importance of sulfation with regard to binding of gp1120/CD4 complexes with CCR5 peptides (Cormier et al., 2000 Proc. Nat. Acad. Sci USA 97:5762-5767). As such, it has been recognized that the disruption of the sulfation of CCR5 may be a treatment for HIV infection and disease processes. In another example, Liu et al. 2008 (Am J Resp Cell Molec Biol 38:738-743) hypothesized that sulfation was a general feature of cytokine receptors and found that at least 10 different cytokine receptors that are involved in asthma and chronic obstructive pulmonary disease (COPD) are sulfated. On this basis, the authors concluded that incorporation of this discovery into the structural design of receptor antagonists might show value in the development of effective drug therapies for asthma, COPD and similar inflammatory lung diseases.
Changes in sulfation patterns have also been found for tumor derived enzymes (Itkonen et al., 2007 FEBS Journal 275:289-301 and a dependency on sulfation has been shown for binding of P-selectin to cancer cells (Ma and Geng 2002 J Immunol 168:1690-1696) and tumorigenesis (Feng et al., 2010 J Vir 84:3351-3361).